Influenza virus is a member of Orthomyxoviridae family (for review, see Murphy and Webster, 1996). There are three subtypes of influenza viruses designated A, B, and C. The influenza virion contains a segmented negative-sense RNA genome. The influenza virion includes the following proteins: hemagglutinin (HA), neuraminidase (NA), matrix (M1), proton ion-channel protein (M2), nucleoprotein (NP), polymerase basic protein 1 (PB1), polymerase basic protein 2 (PB2), polymerase acidic protein (PA), and nonstructural protein 2 (NS2) proteins. The HA, NA, M1, and M2 are membrane associated, whereas NP, PB1, PB2, PA, and NS2 are nucleocapsid associated proteins. The NS1 is the only nonstructural protein not associated with virion particles but specific for influenza-infected cells. The M1 protein is the most abundant protein in influenza particles. The HA and NA proteins are envelope glycoproteins, responsible for virus attachment and penetration of the viral particles into the cell, and the sources of the major immunodominant epitopes for virus neutralization and protective immunity. Both HA and NA proteins are considered the most important components for prophylactic influenza vaccines because they are highly immunogenic.
Influenza virus infection is initiated by the attachment of the virion surface HA protein to a sialic acid-containing cellular receptor (glycoproteins and glycolipids). The NA protein mediates processing of the sialic acid receptor, and virus penetration into the cell depends on HA-dependent receptor-mediated endocytosis. In the acidic confines of internalized endosomes containing an influenza virion, the HA protein undergoes conformational changes that lead to fusion of viral and host cell membranes followed by virus uncoating and M2-mediated release of M1 proteins from nucleocapsid-associated ribonucleoproteins (RNPs), which migrate into the cell nucleus for viral RNA synthesis. Antibodies to HA molecule can prevent virus infection by neutralizing virus infectivity, whereas antibodies to NA proteins mediate their effect on the early steps of viral replication.
To date, all commercially available influenza vaccines for non-pandemic strains in the United States have been propagated in embryonated hen's eggs. Although influenza virus grows well in hen's eggs, production of vaccine is dependent on the availability of eggs. Supplies of eggs must be organized, and strains for vaccine production selected months in advance of the next flu season, limiting the flexibility of this approach, and often resulting in delays and shortages in production and distribution. Unfortunately, some influenza vaccine strains, do not replicate well in embryonated chicken eggs, and have to be isolated by cell culture in a costly and time consuming procedure.
Systems for producing influenza viruses in cell culture have also been developed in recent years (See, e.g., Furminger. Vaccine Production, in Nicholson et al. (eds) Textbook of Influenza pp. 324-332; Merten et al. (1996) Production of influenza virus in cell cultures for vaccine preparation, in Cohen & Shafferman (eds) Novel Strategies in Design and Production of Vaccines pp. 141-151). Typically, these methods involve the infection of suitable immortalized host cells with a selected strain of virus. While eliminating many of the difficulties related to vaccine production in hen's eggs, not all pathogenic strains of influenza grow well and can be produced according to established tissue culture methods. In addition, many strains with desirable characteristics, e.g., attenuation, temperature sensitivity and cold adaptation, suitable for production of live attenuated vaccines, have not been successfully grown in tissue culture using established methods. In addition, live attenuated viruses have not been accepted by the general public due to fears reversion to a virulent virus.
Inactivated influenza A and B virus vaccines are licensed currently as trivalent vaccines for parenteral administration. These trivalent vaccines are produced as monovalent bulk in the allantoic cavity of embryonated chick eggs, purified by rate zonal centrifugation or column chromatography, inactivated with formalin or β-propiolactone, and formulated as a blend of the two strains of type A and the type B strain of influenza viruses in circulation among the human population for a given year. The available commercial influenza vaccines are whole virus (WV) or subvirion (SV; split or purified surface antigen) virus vaccines. The WV vaccine contains intact, inactivated virions. SV vaccines treated with solvents such as tri-n-butyl phosphate (Flu-Shield, Wyeth-Lederle) contain nearly all of the viral structural proteins and some of the viral envelopes. SV vaccines solubilized with Triton X-100 (Fluzone, Sanofi-Aventis; Fluvirin, Novartis) contain aggregates of HA monomers, NA, and NP principally, although residual amounts of other viral structural proteins are present. A live attenuated cold-adapted virus vaccine (FluMist, MedImmune) was granted marketing approval recently by the FDA for commercial usage as an intranasally delivered vaccine indicated for active immunization and the prevention of disease caused by influenza A and B viruses in healthy children and adolescents, 5-17 years of age and healthy adults 18-49 years of age.
Several recombinant products have been developed as recombinant influenza vaccine candidates. These approaches have focused on the expression, production, and purification of influenza virus type A HA and NA proteins, including expression of these proteins using baculovirus infected insect cells (Crawford et al, 1999; Johansson, 1999; Treanor et al., 1996), viral vectors (Pushko et al., 1997; Berglund et al., 1999), and DNA vaccine constructs (Olsen et al., 1997).
Crawford et al. (1999) demonstrated that influenza HA expressed in baculovirus infected insect cells is capable of preventing lethal influenza disease caused by avian H5 and H7 influenza subtypes. At the same time, another group demonstrated that baculovirus-expressed influenza HA and NA proteins induce immune responses in animals superior to those induced by a conventional vaccine (Johansson et al., 1999) Immunogenicity and efficacy of baculovirus-expressed hemagglutinin of equine influenza virus was compared to a homologous DNA vaccine candidate (Olsen et al., 1997). Taken together, the data demonstrated that a high degree of protection against influenza virus challenge can be induced with recombinant HA or NA proteins, using various experimental approaches and in different animal models.
Lakey et al. (1996) showed that a baculovirus-derived influenza HA vaccine was well-tolerated and immunogenic in human volunteers in a Phase I dose escalation safety study. However, results from Phase II studies conducted at several clinical sites in human volunteers vaccinated with several doses of influenza vaccines comprised of HA and/or NA proteins indicated that the recombinant subunit protein vaccines did not elicit protective immunity [G. Smith, Protein Sciences; M. Perdue, USDA, Personal Communications]. These results indicated that conformational epitopes displayed on the surface of HA and NA peplomers of infectious virions were important in the elicitation of neutralizing antibodies and protective immunity.
Regarding the inclusion of other influenza proteins in recombinant influenza vaccine candidates, a number of studies have been carried out, including the experiments involving influenza nucleoprotein, NP, alone or in combination with M1 protein (Ulmer et al., 1993; Ulmer et al., 1998; Zhou et al., 1995; Tsui et al., 1998). These vaccine candidates, which were composed of quasi-invariant inner virion proteins, elicited a broad spectrum immunity that was primarily cellular (both CD4+ and CD8+ memory T cells). These experiments involved the use of the DNA or viral genetic vectors. Relatively large amounts of injected DNA were needed, as results from experiments with lower doses of DNA indicated little or no protection (Chen et al., 1998). Hence, further preclinical and clinical research may be required to evaluate whether such DNA-based approaches involving influenza NP and M1 are safe, effective, and persistent.
Recently, in an attempt to develop more effective vaccines for influenza, particulate proteins were used as carriers of influenza M2 protein epitopes. The rationale for development of an M2-based vaccine was that in animal studies protective immunity against influenza was elicited by M2 proteins (Slepushkin et al., 1995). Neirynck et al. (1999) used a 23-aa long M2 transmembrane domain as an amino terminal fusion partner with the hepatitis B virus core antigen (HBcAg) to expose the M2 epitope(s) on the surface of HBcAg capsid-like particles. However, in spite of the fact that both full-length M2 protein and M2-HBcAg VLP induced detectable antibodies and protection in mice, it was unlikely that future influenza vaccines would be based exclusively on the M2 protein, as the M2 protein was present at low copy number per virion, was weakly antigenic, was unable to elicit antibodies that bound free influenza virions, and was unable to block virus attachment to cell receptors (i.e. virus neutralization).
Since previous research has shown that the surface influenza glycoproteins, HA and NA, are the primary targets for elicitation of protective immunity against influenza virus and that M1 provides a conserved target for cellular immunity to influenza, a new vaccine candidate may include these viral antigens as a protein macromolecular particle, such as virus-like particles (VLPs). Further, the particle with these influenza antigens may display conformational epitopes that elicit neutralizing antibodies to multiple strains of influenza viruses.
Virus-like particles mimic the overall structure of a virus particle without the requirement of containing infectious material. VLPs lack a viral DNA or RNA genome, but retain the three-dimensional structure of an authentic virus. VLPs have the ability to stimulate B-cell mediated responses, CD4 proliferative responses and cytotoxic T lymphocytes responses (see, Schirmbeck et al. (1996) Eur. J. Immunol., 26, 2812-2822). In addition, virus-like particles induce MHC class I-restricted T-cell responses.
Several studies have demonstrated that recombinant influenza proteins could self-assemble into VLPs in cell culture using mammalian expression plasmids or baculovirus vectors (Gomez-Puertas et al., 1999; Neumann et al., 2000; Latham and Galarza, 2001). Gomez-Puertas et al. (1999) demonstrated that efficient formation of influenza VLP depends on the expression levels of viral proteins. Neumann et al. (2000) established a mammalian expression plasmid-based system for generating infectious influenza virus-like particles entirely from cloned cDNAs. Latham and Galarza (2001) reported the formation of influenza VLPs in insect cells infected with recombinant baculovirus co-expressing human influenza virus HA, NA, M1, and M2 genes. These studies demonstrated that influenza virion proteins may self-assemble upon co-expression in eukaryotic cells.
However, one problem associated with the use of the M1 protein from human strains of influenza virus is that they are poor proteins for efficient VLP formation. Indeed, the present inventors have found that the use of the M1 protein from human seasonal strains results in low quantities of VLPs that are not sufficient for commercial VLP production. Suprisingly, the present inventors found that M1 proteins derived from avian strains of influenza virus are much more favorable proteins for efficient VLP production. This increased efficiency was found to be mediated in part by a single amino acid difference in the M1 protein (an R to K substitution at position 101 of the M1 protein). This mutation was found almost exclusively in avian M1 proteins. Importantly, the present inventors have found that in order to produce recoverable levels of VLPs sufficient for vaccine production, it is necessary to use M1 proteins, such as avian M1 proteins, harboring the K101 amino acid residue. Accordingly, the knowledge that increased formation and recovery of VLPs using M1 proteins containing this amino acid substitution is critical to vaccine development.